Current human genome databases for public single nucleotide polymorphisms (SNPs) still contain a substantial fraction of false entries. The main reasons for errors include sequencing or assembly errors, paralogous sequence-, and private variants. In the course of our studies on the Y chromosome, we established a set of internal laboratory guidelines for reliably identifying false SNP entries in databases.